A guide to hypothesis testing
Click here to load reader
-
Upload
modelos-econometricos -
Category
Technology
-
view
1.881 -
download
2
description
Transcript of A guide to hypothesis testing
ECON 351* -- A Guide to Hypothesis Testing M.G. Abbott
A Guide to Hypothesis Testing in Linear Regression Models Tests of One Coefficient Restriction: One Restriction on One Coefficient H0 specifies only one equality restriction on one coefficient.
Two-Tail Tests of One Restriction on One Coefficient • Example: H0: βj = bj versus H1: βj ≠ bj where bj is a specified constant. • Use: either a two-tail t-test or an F-test. • Test Statistics:
tse
t N kjj j
j
( $ )$
$( $ )~ [ ]β
β β
β=
−− ⇒ sample value =
)ˆ(esbˆ
)ˆ(j
jjj0
β
−β=βt .
( )
FVar
F N kj
j j
j
( $ )$
$ ( $ )~ [ ,β
β β
β=
−−
2
1 ] ⇒ sample value = ( )
)ˆ(raVbˆ
)ˆ(j
2
jjj0 β
−β=βF .
Note: [ ] or t Fj j( $ ) ( $ )β β2= )ˆ(F)ˆ(t jj β=β and [ ] ]kN,1[F]kN[t 2
2/ −=− αα .
• Decision Rules:
Reject H0 if ]kN[tt 2/0 −> α or two-tail p-value for ( ) α<>= 00 ttPrt ; or p-value for ]kN,1[FF0 −> α α<>= )FFPr(F 00 .
Retain H0 if ]kN[tt 2/0 −≤ α or two-tail p-value for ( ) α≥>= 00 ttPrt ; or p-value for ]kN,1[FF0 −≤ α α≥>= )FFPr(F 00 .
ECON 351*: Guide to Hypothesis Testing Page 1 of 5 351guide.doc
ECON 351* -- A Guide to Hypothesis Testing M.G. Abbott
One-Tail Tests of One Restriction on One Coefficient • Examples: H0: βj = bj (or βj ≥ bj) versus H1: βj < bj a left-tail test H0: βj = bj (or βj ≤ bj) versus H1: βj > bj a right-tail test • Use: a one-tail t-test. • Test Statistic:
tse
t N kjj j
j
( $ )$
$( $ )~ [ ]β
β β
β=
−− ⇒ sample value =
)ˆ(es
bˆ)ˆ(t
j
jjj0
β
−β=β .
• Decision Rules -- left-tail t-test:
Reject H0 if or one-tail p-value for ]kN[tt 0 −−< α α<<= )ttPr(t 00 ; Retain H0 if or one-tail p-value for ]kN[tt 0 −−≥ α α≥<= )ttPr(t 00 .
• Decision Rules -- right-tail t-test:
Reject H0 if or one-tail p-value for ]kN[tt 0 −> α α<>= )ttPr(t 00 ; Retain H0 if or one-tail p-value for ]kN[tt 0 −≤ α α≥>= )ttPr(t 00 .
ECON 351*: Guide to Hypothesis Testing Page 2 of 5 351guide.doc
ECON 351* -- A Guide to Hypothesis Testing M.G. Abbott
Tests of One Linear Restriction on Two or More Coefficients H0 specifies only one linear restriction on two or more regression coefficients.
Two-Tail Tests of One Linear Restriction on Two Coefficients • Example: H0: cjβj + chβh = c0 versus H1: cjβj + chβh ≠ c0. • Use: either a two-tail t-test or an F-test. • Test Statistics:
( ) ]kN[t]kN[t~)ˆcˆc(es
)cc()ˆcˆc(ˆcˆct 1hhjj
hhjjhhjjhhjj −=−
β+β
β+β−β+β=β+β
where: )ˆcˆc(raV)ˆcˆc(e hhjjhhjj β+β=β+βs
. )ˆ,ˆ(voCcc2)ˆ(raVc)ˆ(raVc)ˆcˆc(raV hjhjh2hj
2jhhjj ββ+β+β=β+β
sample value = ( ))ˆcˆc(esc)ˆcˆc(ˆcˆct
hhjj
0hhjjhhjj0 β+β
−β+β=β+β .
( ) [ ]]kN,1[F]kN,1[F~
)ˆcˆc(raV)cc()ˆcˆc(ˆcˆcF 1
hhjj
2
hhjjhhjjhhjj −=−
β+β
β+β−β+β=β+β
where: . )ˆ,ˆ(voCcc2)ˆ(raVc)ˆ(raVc)ˆcˆc(raV hjhjh2hj
2jhhjj ββ+β+β=β+β
sample value = ( ) [ ])ˆcˆc(raV
c)ˆcˆc(ˆcˆcFhhjj
2
0hhjjhhjj0 β+β
−β+β=β+β
• Decision Rules:
Reject H0 if ]kN[tt 2/0 −> α or two-tail p-value for ( ) α<>= 00 ttPrt ; or p-value for ]kN,1[FF0 −> α α<>= )FFPr(F 00 .
Retain H0 if ]kN[tt 2/0 −≤ α or two-tail p-value for ( ) α≥>= 00 ttPrt ; ] or p-value for kN,1[FF0 −≤ α α≥>= )FFPr(F 00 .
ECON 351*: Guide to Hypothesis Testing Page 3 of 5 351guide.doc
ECON 351* -- A Guide to Hypothesis Testing M.G. Abbott
One-Tail Tests of One Linear Restriction on Two Coefficients • Examples: H0: cjβj + chβh = c0 vs. H1: cjβj + chβh < c0 a left-tail test H0: cjβj + chβh = c0 vs. H1: cjβj + chβh > c0 a right-tail test • Use: a one-tail t-test. • Test Statistic:
( ) ]kN[t]kN[t~)ˆcˆc(es
)cc()ˆcˆc(ˆcˆct 1hhjj
hhjjhhjjhhjj −=−
β+β
β+β−β+β=β+β
where )ˆcˆc(raV)ˆcˆc(e hhjjhhjj β+β=β+βs
)ˆ,ˆ(voCcc2)ˆ(raVc)ˆ(raVc)ˆcˆc(raV hjhjh2hj
2jhhjj ββ+β+β=β+β .
sample value = ( ))ˆcˆc(esc)ˆcˆc(ˆcˆct
hhjj
0hhjjhhjj0 β+β
−β+β=β+β .
• Decision Rules -- left-tail t-test:
Reject H0 if or one-tail p-value for ]kN[tt 0 −−< α α<<= )ttPr(t 00 ; Retain H0 if or one-tail p-value for ]kN[tt 0 −−≥ α α≥<= )ttPr(t 00 .
• Decision Rules -- right-tail t-test:
Reject H0 if or one-tail p-value for ]kN[tt 0 −> α α<>= )ttPr(t 00 ;
Retain H0 if or one-tail p-value for ]kN[tt 0 −≤ α α≥>= )ttPr(t 00 .
ECON 351*: Guide to Hypothesis Testing Page 4 of 5 351guide.doc
ECON 351* -- A Guide to Hypothesis Testing M.G. Abbott
ECON 351*: Guide to Hypothesis Testing Page 5 of 5 351guide.doc
Tests of Two or More Linear Coefficient Restrictions H0 specifies two or more linear coefficient restrictions. • Example: H0: β2 = β4 and β3 = β5
H1: β2 ≠ β4 and/or β3 ≠ β5 • Use: a general F-test; only an F-test can be used to test jointly two or more
coefficient restrictions. • Test Statistics: Either of the following two general F-statistics.
)kN(RSS
)kk()RSSRSS(dfRSS
)dfdf()RSSRSS(F
1
010
11
1010
−−−
=−−
= .
)kN()R1(
)kk()RR(df)R1(
)dfdf()RR(F 2
U
02R
2U
12U
102R
2U
−−−−
=−
−−= .
Null distribution: ]kN,kk[F]df,dfdf[F~F 0110 −−=− .
• Sample value of F-statistic under H0 = F0. • Decision Rules:
Reject H0 if ]kN,kk[F]df,dfdf[FF 01100 −−=−> αα or p-value for α<>= )FFPr(F 00 .
Retain H0 if ]kN,kk[F]df,dfdf[FF 01100 −−=−≤ αα or
p-value for α≥>= )FFPr(F 00 .